Physical mechanisms and hardware apparatus for generative display content
In some embodiments, generative systems are physical mechanics and hardware apparatus to generate display content and include an input receiver module, a graphics processor module, and an output module. In some embodiments, the input receive module converts or splits a video input signal to provide a video input stream for the graphical processing module, which includes one or more processing units. The graphics processor module extracts both the graphical input content and the metadata of the input stream, operates on said information, and outputs the result to the output module, which emits an output signal for display on a display system. In some embodiments, the graphics processor module uses a generative function or a computer vision function. In some embodiments generative system is a standalone device to mediate and infer visual content between a computational system and a display system.
1 . A generative system comprising a housing, and within the housing:
an input stream generator to receive a visual input stream from a main computer;
a function executor coupled to the input stream generator (i) to identify a modification of the visual input stream based on a plurality of user actions via a user input device, (ii) to calculate a property of the visual input stream based on the modification, and (iii) to automatically produce a generative content based on the property; and
a display controller to output the generative content to a display system, wherein the housing is external to the main computer and to the display system.
2 . The generative system of claim 1 , wherein the function executor operates on the visual input stream with a function chosen from a set consisting of a generative artificial intelligence (AI) function, a computer vision function, a geometric warping function, and combinations thereof.
3 . The generative system of claim 1 , wherein the input stream generator further receives metadata about the visual input stream, and the generative content depends in part on the metadata.
4 . The generative system of claim 1 , wherein the visual input stream is a first visual input stream among a plurality of visual input streams received by the input stream generator.
5 . The generative system of claim 1 , wherein the function executor uses a cloud resource.
6 . The generative system of claim 1 , wherein the display system is a free-standing virtual display system, the generative content shown in an image having at least one monocular depth.
7 . A generative system comprising a housing, and within the housing:
an input stream generator to receive a visual input stream from a main computer;
a function executor coupled to the input stream generator (i) to identify a modification of the visual input stream based on a plurality of user actions via a user input device, (ii) to calculate a property of the visual input stream based on the modification, and (iii) to automatically produce a generative content based on the property; and
a display controller to output the generative content to a display system, wherein the housing is external to the main computer and to the display system, and wherein
the display system is a multilayer display system;
the visual input stream is directly shown on a first focal plane of the multilayer display system; and
the generative content is shown on a second focal plane of the multilayer display system.
8 . The generative system of claim 7 , further comprising at least one non-transitory computer-readable storage medium storing computer-executable code that causes:
the input stream generator to receive the visual input stream from the main computer;
the function executor (i) to identify the modification of the visual input stream based on the plurality of user actions via the user input device, (ii) to calculate the property of the visual input stream based on the modification, and (iii) to automatically produce the generative content based on the property; and
the display controller to output the generative content to the display system.
9 . The generative system of claim 1 , wherein the display system is a television, a phone, a watch, a headset, a laptop screen, a smart device, a computer monitor, or a tablet.
10 . The generative system of claim 1 , wherein the function executor comprises at least one graphics processor unit (GPU), and the function executor creates the generative content in real time.
11 . The generative system of claim 1 , further comprising:
an input port to couple the visual input stream from an input signal cable to the input stream generator; and
an output port to couple the generative content to an output signal cable.
12 . The generative system of claim 1 , wherein the generative content includes at least a portion of the visual input stream.
13 . The generative system of claim 1 , wherein the modification is impacted by metadata about a user interface (UI).
14 . The generative system of claim 13 , wherein the metadata includes a window location, a window layout, a cursor location, an activated window, or at least one coordinate of an on-screen graphic.
15 . The generative system of claim 1 , wherein the property is a mathematical correlation between a first frame of the visual input stream and a second frame of the visual input stream.
16 . The generative system of claim 1 , wherein the generative display content comprises a plurality of future possible tasks.
17 . The generative system of claim 16 , wherein the generative system further instructs the main computer to automatically execute at least one of the plurality of future possible tasks.
18 . The generative system of claim 1 , further comprising at least one non-transitory computer-readable storage medium storing computer-executable code that causes:
the input stream generator to receive the visual input stream from the main computer;
the function executor (i) to identify the modification of the visual input stream based on the plurality of user actions via the user input device, (ii) to calculate the property of the visual input stream based on the modification, and (iii) to automatically produce the generative content based on the property; and
the display controller to output the generative content to the display system.
19 . A generative system comprising:
an input stream generator to receive a visual input stream from a source;
a function executor coupled to the input stream generator to operate on the visual input stream and to create generative content derived from the visual input stream; and
a display controller to output the generated content to a display system,
wherein the generative system is a standalone device to interface between the source and the display system, and wherein
the generative content is determined at least in part by a metadata of the visual input stream caused by a user action via a user input device.
20 . The generative system of claim 19 , wherein the function executor uses a cloud resource.
21 . The generative system of claim 19 , wherein the generative system is a portable dongle.
22 . The generative system of claim 19 , wherein the function executor operates on the visual input stream with a function chosen from a set consisting of a generative artificial intelligence (AI) function, a computer vision function, a geometric warping function, and combinations thereof.
23 . The generative system of claim 19 , wherein the function executor creates the generative content in real time.
24 . The generative system of claim 19 , further comprising at least one non-transitory computer-readable storage medium storing computer-executable code that causes:
the input stream generator to receive the visual input stream from the source;
the function executor to operate on the visual input stream and to create the generative content derived from the visual input stream; and
the display controller to output the generated content to the display system.